On this page
The right source depends on the question you are trying to answer. A company filing is the starting point for reported performance, a proxy statement helps explain governance and incentives, and industry data tests whether a company trend is shared by peers. For CFA® Level I, the skill is to select the evidence that answers the question and judge what that evidence can actually prove.
Start with the question, then select the document
The CFA Institute introduction to financial statement analysis identifies filings, notes, management commentary, audit reports, issuer communications and outside research as distinct inputs. Treat that list as a map of possible evidence, not as an instruction to read every document from beginning to end. If you need to know whether revenue was recognised, inspect the statements and accounting notes. If you need to know why management believes demand improved, read its commentary and then look for corroboration. The two questions call for different degrees of confidence.
A useful first pass separates three layers. Reported figures and accounting policies tell you what the issuer recorded under its reporting framework. Management discussion explains how the issuer interprets those figures and known trends. Outside data provides context that the issuer may not control. None of these layers replaces the others. An audited balance sheet does not by itself reveal a new industry demand shock, while a trade association chart cannot establish the precise accounting treatment of one company.
The SEC guide to Form 10-K describes the business section, risk factors, management discussion, financial statements, notes and auditor report. It also says the SEC does not prepare or vouch for the company filing. That distinction matters in an exam answer: a regulatory filing is a required disclosure channel, but its presence on a regulator website is not a guarantee that every management assertion is true. Look at which part is audited and which part expresses a view.
Match common analyst questions to their best starting points
For a historical profit or cash-flow question, begin with the annual filing and its notes. The main statements provide amounts, while the notes explain policies, estimates and changes that can make apparently similar figures behave differently. Compare periods on a consistent basis before drawing a trend. A large movement in receivables alongside revenue growth, for example, raises a question about collection timing; it does not establish a reporting failure. The notes and subsequent filings help test competing explanations.
For a more recent operating trend, use the interim report. It is usually closer to the period of interest than the annual filing, but it may be less comprehensive and should not be casually treated as another full annual audit. Read the period heading and the basis of comparison. A quarterly amount, a year-to-date amount and a trailing-year amount answer different questions even if all appear in one document. Seasonality can make a sequential comparison misleading, so a candidate should say exactly what is being compared.
For management incentives or shareholder votes, seek the proxy materials. Compensation design can reveal which outcomes are rewarded, and governance disclosures can show potential conflicts or oversight arrangements. An incentive is a reason to investigate, not proof that a manager manipulated a reported number. Likewise, a board biography says something about stated experience, but its existence does not establish the effectiveness of oversight. Keep the inference proportionate to the evidence.
For a company narrative about a sharp move in sales, read the earnings release or call with a verification question in mind. The narrative can identify a claimed driver faster than the next filing, yet it is issuer-controlled. A claim that higher sales reflect unit growth is stronger if volume disclosures or an independently collected industry series point the same way. If the increase could instead reflect prices, acquisitions or currency, request the bridge between those drivers before inferring stronger underlying demand.
For industry conditions, use government releases, trade data and peer disclosures, while checking definitions and period coverage. The CFA Institute reading names third-party research and government economic information as supplementary sources. An industry series might cover a different geography, product mix or reporting interval from the company. It can challenge an issuer claim, but a mismatch alone does not establish that the issuer is wrong.
A worked source-selection exercise
Suppose a fictional manufacturer reports annual revenue of 100 units of currency and, in its next interim report, revenue of 30 for a quarter. A management presentation says the recent increase reflects stronger demand. The candidate also sees an industry report showing flat total units sold in the relevant market. The task is to assess whether the company has gained market share, rather than to calculate a target price. Begin by checking whether the annual 100 and quarterly 30 cover comparable product lines and accounting policies. The quarterly amount is not automatically 30 per cent of the annual amount in an economically meaningful sense because of seasonality and different periods.
Next, separate sales value from units. If the fictional quarter contains 10 units sold at a price of 3, the revenue is 30. If the previous comparable quarter contained 10 units sold at a price of 2, its revenue was 20. Revenue rose by 10, or 50 per cent, yet units did not rise. The arithmetic permits a price-driven explanation; it does not support a volume-driven share gain. The company filing or release might disclose the price-volume bridge, while the industry report can tell you whether the whole market experienced a similar price change.
Now consider a different invented comparison: the company sold 12 units at the same price of 2, giving revenue of 24, while the market remained flat in units. This pattern could be consistent with share gain if company and market unit definitions match and the periods align. It still requires checks for acquisitions, channel inventory and differences between shipments and final sales. The conclusion should say “consistent with” rather than “proves”. This phrasing reflects the actual evidential gap rather than excessive caution for its own sake.
Finally, if management cites an adjusted profit figure, ask which costs were excluded and compare with the most directly comparable reported measure. The SEC non-GAAP guidance warns that labels and adjustments can mislead and that such measures are not necessarily comparable across companies. The issue is not that all adjusted measures are useless. The issue is that a candidate must understand the reconciliation before treating a change in the adjusted number as an economic improvement.
How to weigh conflicting evidence
A sensible hierarchy begins with the decision being tested. When a claim concerns recognised revenue, the financial statements, policies and notes carry more direct weight than a call summary. When it concerns the explanation for revenue growth, the filing may establish the amount while outside volume data tests the explanation. When it concerns future demand, neither a past audited amount nor an optimistic call can settle the matter alone. Give each source a job and identify the uncertainty left after it has done that job.
Check dates, definitions and assurance before comparing. An audit opinion on annual financial statements is not an audit of every slide in an investor presentation. An interim report may contain reviewed financial information without the same assurance as the year-end statements. A proxy disclosure about pay can identify incentives but is not an outcome measure. A survey or database can be useful for context while still carrying sampling, classification or transcription error. State these limits in the answer rather than calling all outside material “unreliable”.
If sources disagree, look for a definition or timing difference first. A reported sales figure may be consolidated, while an industry series covers one region. A company may present constant-currency growth while a filing gives reported growth. A call may refer to orders while the income statement records recognised sales. Resolve those differences before concluding that the data conflict. The candidate earns more by explaining a plausible reconciliation and the missing check than by selecting whichever figure sounds more current.
The mistakes to avoid in a Level I answer
The first mistake is treating “official filing” as synonymous with “audited claim”. A Form 10-K contains audited statements, but it also contains management discussion. The second is using an earnings call to answer a question about an accounting policy when the note is the direct source. The third is reading a proxy as a verdict on management character rather than evidence of incentives and governance. The fourth is comparing peer ratios or industry statistics without aligning periods and definitions.
A compact exam method is to write the information need in one line, name the source that directly addresses it, and add one sentence on its limit. That structure works even when several sources are plausible. For related context, see the CFA programme page and Omni methodology. Use the source-selection question itself as the test: what can this document establish, and what would still need checking?
Sources
- Introduction to Financial Statement Analysis | CFA Institute · cfainstitute.org · Retrieved
- How to Read a 10-K | SEC · sec.gov · Retrieved
- Non-GAAP Financial Measures | SEC · sec.gov · Retrieved



